Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Let the king capture or generate legal moves

Author: Michel Langeveld

Date: 10:56:24 11/20/04

Go up one level in this thread


On November 20, 2004 at 12:05:16, Reinhard Scharnagl wrote:

>Hi Michel,
>
>I think it does not matter, where you are investing the effort to check the
>legality of a move. You can do it by making a move or by testing its before.
>I test it before I move, thus the last ply is not moved. I think it has not
>that influence you supposed, because if I would make a unecessare move in
>the last ply I could have ommitted my legality checking parts of the program.
>
>But if you would think I am interested in fast results by using some tricks:
>
>a) I would not have made a detailed move type statistic (about 20% of run-time)
>
>b) I would have posted following results:

:-) Yes, very nice made with hashing of perft values.
Do you use a 64bit hash-id of the position or the complete position as hashid?
The former is not so safe to use. But you probably know that.

>FEN: rnbqckabnr/pppppppppp/0/0/0/0/PPPPPPPPPP/RNBQCKABNR w KQkq - 0 1
>
>  +-*--b--c--d--e--*--g--h--i--*-+ MS Vis. Studio C++ Vers. 13.10
>8 |[r][n][b][q][c][k][a][b][n][r]| (Compilation: Nov 17 2004)
>7 |[p][p][p][p][p][p][p][p][p][p]|
>6 |   :::   :::   :::   :::   :::| Perft Testseries
>5 |:::   :::   :::   :::   :::   | (With TT Caching 256.0 MB / 4-fold)
>4 |   :::   :::   :::   :::   :::| TT Access Success 53.1%
>3 |:::   :::   :::   :::   :::   |
>2 |<P><P><P><P><P><P><P><P><P><P>| Smirf Test No.: 07
>1 |<R><N><B><Q><C><K><A><B><N><R>|
>=>+-*--b--c--d--e--*--g--h--i--*-+ Break Time 37.5 Sec.
>
>Ply           Nodes          all (x)          all (+)         Castl.    Sec.
>----------------------------------------------------------------------------
>1                28                0                0              0       0
>2               784                0                0              0       0
>3             25283              387              120              0       0
>4            808984            14812             4019              0       0
>5          28946187           907953           304360              0     0.5
>6        1025229212         36740962         11052055              0     9.1
>7       39532257395       1838308392        621202264            120   155.1
>----------------------------------------------------------------------------
>
>Regards, Reinhard.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.